The term “code” I’m talking here is not the concept in programming, meaning that we use programming language to write scripts or programs. In this context, code is one step in the procedure of qualitative research. I’m not going to spend too much length explaining the specific coding process here. Rather, I organized some tips from my own coding experiences, which I think may be useful for new coders to avoid some pitfalls, as well as for myself to pay attention to in my future research.
A brief introduction to Qualitative Research and code
Qualitative research, different from quantitative research which emphasize objective measurements and the statistical analysis of data collected through polls, questionnaires, and surveys, or other statistical data using computational techniques, is primarily exploratory research. It is used to gain an understanding of underlying issues. After preparing for qualitative research, the general steps left are: 1. Collect data; 2. Analyze data; and 3. Write report.
Collect data: In qualitative research, we can use interview, field trip, requesting data by API, etc. to collect data. The data can be interview transcripts, documents, artifacts, and so on. In the Celebrities and Fans project, we asked pre-designed and follow-up questions to get data from our interviewees, and recorded these interviews for further analysis. In the NPO project, we wrote scripts to fetch Facebook timeline feed though API.
Analyze data: For most time, this step means textual analysis. In the Celebrities and Fans project, we analyzed interview transcripts; and in the NPO project, we analyzed the Facebook posts content from NPOs’ timeline. Coding is one way to analyze data. A code in qualitative method is “most often a word or short phrase that symbolically assigns a summative, salient, essence-capturing, and/or evocative attribute for a portion of language-based or visual data (Saldaña, 2009, P3).”
Write report: Document your analyzing results and findings. Make sure your results contribute to your early research.
My experience of coding
I firstly studies the method by conducting the “Celebrities and Fans” research. I coded 17 transcripts from fans of The Walking Dead and 9 transcripts from fans of House of Cards. And then I practiced this method again through coding 600 NPOs Facebook posts. I personally think that my second practice is much better than my first one. I’ve learnt a lot through doing the research. And I also reviewed and reflected on my pratice. There were some things that I did good, and some things that I overlooked. Here are some tips that I’ve got from my experiences.
Keep your goal in mind.
Always keep your goal in mind when you are coding. Coding is basically a kind of classification based on the criteria that created by yourself. And when you are coding, you are proposing some criteria that categorizes these entries. The criteria are very crucial and it can affect your results a lot. For instance, a post from one NPO, Aids Foundation of Chicago, is about some simple ways to take precautions against aids infection. This post can be easily classified as “aids.” However, this “aids” code is not a good one. What we are trying to do through the analysis is to figure out how NPOs use social media. Our objects are general NPOs in Chicago which may focus on various fields besides aids. Therefore, this code does not seem to fit for all NPOs. As we read more posts, we may understand that this kind of post is to educate people and distribute basic knowledge in their specific field. So we eventually used the code “professional knowledge” to describe this kind of posts.
Know the context.
This tip comes along with the first one. Knowing more about the context means to have some basic knowledge not only about your own research goal, but also about your research objects and their goals. For instance, one post about a rap singer’s new released album might look irrelevant to our research theme of civic engagement at the beginning. But when we looked into the NPO who created this post, we realized that this NPO contributed to address public attention to African American’s growth and achievement. And the singer mentioned in the post was a young African American singer. Therefore, this post was not some irrelevant, gossip news. Instead, it’s way of drawing awareness to the African American group through addressing popular culture. Additionally, knowing the context is also essential to design interview questions.
Keep code concrete.
Code is only some word(s) or phrases. And we later on will create categories and themes by summarizing them. If one code contains too many more entries than any other entries, it may be reasonable to consider if it’s too broad. If one code is too broad, the whole result can be lack in variation because other codes have very little entries. And the code process will make no sense. Therefore, when a code is too broad, it may need to be split. For instance, when we interviewed fans, we asked a lot of how they use social media. At the beginning, we only used the code “social media” to summarize these kinds of sentences. Sooner, this code became so big and those sentences were very different: some were about how they got information from social media, some were about their strategies of using various social media sites, some were about the community they interacted with on social media, etc. Therefore, we then created sub-codes to split this code.
Explain your code with others.
One thing that I kept doing during the coding process was to repeatedly check if the code was appropriate to summarize patterns of behaviors. And I also wanted to make sure that other researchers can understand my code. I wanted my codes to be readable and have a high legibility. This is especially important when you are doing coding with one or more coders together. It is useful to explain and discuss the name and definition of each code, and if possible, write down the definition and example of each code. It will save a lot of communication and time.
Take notes after coding.
Don’t not start to write too late! I did it late in my first practice in the fans project. When your memory is still fresh, write down your thoughts and ideas is a very good habit. You don’t have to get back and go through everything again when you start to write the report. Especially for interviews and the research that you have to conduct recursively, these swift feedback from coding can give you useful suggestions on how to modify your future interview questions.
Coding takes a long time. But if you have a definite and clear goal about what you expect to get through analyzing these data, your coding process will be effective and faster. Therefore, I’m suggesting that anyone who do coding, they need to at least have some basic knowledge about the field they are contributing to and the possible goal they want to achieve. And if possible, it’s good to have two coders to code and discuss during their practices. It is also a good habit to take notes just after every time coding process, so that that temporary memory will not fade as time goes by.
Saldana, J. (2012). The Coding Manual for Qualitative Researchers. SAGE.